Merge branch 'release/v0.2.0'

v0.1.9

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2020 taizan-hokuto
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

pytchat/__init__.py

@@ -2,7 +2,7 @@
 pytchat is a lightweight python library to browse youtube livechat without Selenium or BeautifulSoup.
 """
 __copyright__    = 'Copyright (C) 2019 taizan-hokuto'
-__version__      = '0.1.0'
+__version__      = '0.2.0'
 __license__      = 'MIT'
 __author__       = 'taizan-hokuto'
 __author_email__ = '55448286+taizan-hokuto@users.noreply.github.com'

pytchat/cli/__init__.py

@@ -1,11 +1,17 @@
 import argparse
+
+import os
+import sys
+import signal
+from json.decoder import JSONDecodeError
 from pathlib import Path
-from pytchat.util.extract_video_id import extract_video_id
 from .arguments import Arguments
-from .. exceptions import InvalidVideoIdException, NoContents
+from .. exceptions import InvalidVideoIdException, NoContents, PatternUnmatchError
 from .. processors.html_archiver import HTMLArchiver
 from .. tool.extract.extractor import Extractor
 from .. tool.videoinfo import VideoInfo
+from .. util.extract_video_id import extract_video_id
+from .. import util
 from .. import __version__
 
 '''
@@ -20,47 +26,101 @@ https://github.com/PetterKraabol/Twitch-Chat-Downloader
 def main():
     # Arguments
     parser = argparse.ArgumentParser(description=f'pytchat v{__version__}')
-    # parser.add_argument('VideoID_or_URL', type=str, default='__NONE__',nargs='?',
-    #                     help='Video ID, or URL that includes id.\n'
-    #                     'If ID starts with a hyphen (-), enclose the ID in square brackets.')
     parser.add_argument('-v', f'--{Arguments.Name.VIDEO_IDS}', type=str,
-                        help='Video ID (or URL that includes Video ID). You can specify multiple video IDs by separating them with commas without spaces.\n'
+                        help='Video ID (or URL that includes Video ID). You can specify multiple video IDs by '
+                        'separating them with commas without spaces.\n'
                         'If ID starts with a hyphen (-), enclose the ID in square brackets.')
     parser.add_argument('-o', f'--{Arguments.Name.OUTPUT}', type=str,
                         help='Output directory (end with "/"). default="./"', default='./')
+    parser.add_argument(f'--{Arguments.Name.PBAR}', action='store_true',
+                        help='Display rich progress bar')
+    parser.add_argument(f'--{Arguments.Name.SAVE_ERROR_DATA}', action='store_true',
+                        help='Save error data when error occurs(".dat" file)')
     parser.add_argument(f'--{Arguments.Name.VERSION}', action='store_true',
                         help='Show version')
     Arguments(parser.parse_args().__dict__)
+
+    if Arguments().pbar:
+        from .progressbar_rich import ProgressBar
+    else:
+        from .progressbar_simple import ProgressBar
     if Arguments().print_version:
         print(f'pytchat v{__version__}     © 2019 taizan-hokuto')
         return
 
     # Extractor
-    if Arguments().video_ids:
+    if not Arguments().video_ids:
-        for video_id in Arguments().video_ids:
+        parser.print_help()
-            if '[' in video_id:
-                video_id = video_id.replace('[', '').replace(']', '')
-            try:
-                info = VideoInfo(video_id)
-                print(f"Extracting...\n"
-                      f" video_id: {video_id}\n"
-                      f" channel:  {info.get_channel_name()}\n"
-                      f" title:    {info.get_title()}")
-                path = Path(Arguments().output + video_id + '.html')
-                print(f" output path: {path.resolve()}")
-                Extractor(video_id,
-                          processor=HTMLArchiver(
-                              Arguments().output + video_id + '.html'),
-                          callback=_disp_progress
-                          ).extract()
-                print("\nExtraction end.\n")
-            except InvalidVideoIdException:
-                print("Invalid Video ID or URL:", video_id)
-            except (TypeError, NoContents) as e:
-                print(e)
         return
-    parser.print_help()
+    for video_id in Arguments().video_ids:
+        if '[' in video_id:
+            video_id = video_id.replace('[', '').replace(']', '')
+        try:
+            video_id = extract_video_id(video_id)
+            if os.path.exists(Arguments().output):
+                path = Path(Arguments().output + video_id + '.html')
+            else:
+                raise FileNotFoundError
+            info = VideoInfo(video_id)
+            print(f"Extracting...\n"
+                  f" video_id: {video_id}\n"
+                  f" channel:  {info.get_channel_name()}\n"
+                  f" title:    {info.get_title()}")
+
+            print(f" output path: {path.resolve()}")
+            duration = info.get_duration()
+            pbar = ProgressBar(total=(duration * 1000), status="Extracting")
+            ex = Extractor(video_id,                  
+                    callback=pbar._disp,
+                    div=10)
+            signal.signal(signal.SIGINT, (lambda a, b: cancel(ex, pbar)))
+            data = ex.extract()
+            if data == []:
+                return False
+            if Arguments().pbar:
+                pbar.reset("#", "=", total=len(data), status="Rendering  ")
+            else:
+                pbar.reset("=", "", total=len(data), status="Rendering  ")
+            processor = HTMLArchiver(Arguments().output + video_id + '.html', callback=pbar._disp)
+            processor.process(
+                [{'video_id': None,
+                'timeout': 1,
+                'chatdata': (action["replayChatItemAction"]["actions"][0] for action in data)}]
+            )
+            processor.finalize()
+            if Arguments().pbar:
+                pbar.reset('#', '#', status='Completed   ')
+                pbar.close()
+            else:
+                pbar.close()
+                print("\nCompleted")
+            
+            print()
+            if pbar.is_cancelled():
+                print("\nThe extraction process has been discontinued.\n")
+                return False
+            return True
+
+        except InvalidVideoIdException:
+            print("Invalid Video ID or URL:", video_id)
+        except NoContents as e:
+            print(e)
+        except FileNotFoundError:
+            print("The specified directory does not exist.:{}".format(Arguments().output))
+        except JSONDecodeError as e:
+            print(e.msg)
+            print("Cannot parse video information.:{}".format(video_id))
+            if Arguments().save_error_data:
+                util.save(e.doc, "ERR_JSON_DECODE", ".dat")
+        except PatternUnmatchError as e:
+            print(e.msg)
+            print("Cannot parse video information.:{}".format(video_id))
+            if Arguments().save_error_data:
+                util.save(e.doc, "ERR_PATTERN_UNMATCH", ".dat")
+
+    return
 
 
-def _disp_progress(a, b):
+def cancel(ex, pbar):
-    print('.', end="", flush=True)
+    ex.cancel()
+    pbar.cancel()

pytchat/cli/arguments.py

@@ -18,6 +18,8 @@ class Arguments(metaclass=Singleton):
         VERSION: str = 'version'
         OUTPUT: str = 'output_dir'
         VIDEO_IDS: str = 'video_id'
+        SAVE_ERROR_DATA: bool = 'save_error_data'
+        PBAR: bool ='pbar'
 
     def __init__(self,
                  arguments: Optional[Dict[str, Union[str, bool, int]]] = None):
@@ -34,10 +36,9 @@ class Arguments(metaclass=Singleton):
         self.print_version: bool = arguments[Arguments.Name.VERSION]
         self.output: str = arguments[Arguments.Name.OUTPUT]
         self.video_ids: List[int] = []
+        self.save_error_data: bool = arguments[Arguments.Name.SAVE_ERROR_DATA]
+        self.pbar: bool = arguments[Arguments.Name.PBAR]
         # Videos
         if arguments[Arguments.Name.VIDEO_IDS]:
             self.video_ids = [video_id
                               for video_id in arguments[Arguments.Name.VIDEO_IDS].split(',')]
-
-
-

pytchat/cli/progressbar_rich.py

@@ -0,0 +1,52 @@
+'''
+This code for this progress bar is based on
+vladignatyev/progress.py
+https://gist.github.com/vladignatyev/06860ec2040cb497f0f3
+(MIT License)
+'''
+import sys
+
+
+class ProgressBar:
+    def __init__(self, total, status):
+        self._bar_len = 60
+        self._cancelled = False
+        self.reset(total=total, status=status)
+        self._blinker = 0
+
+    def reset(self, symbol_done="=", symbol_space=" ", total=100, status=''):
+        self._symbol_done = symbol_done
+        self._symbol_space = symbol_space
+        self._total = total
+        self._status = status
+        self._count = 0
+
+    def _disp(self, _, fetched):
+        self._progress(fetched, self._total)
+
+    def _progress(self, fillin, total):
+        if total == 0 or self._cancelled:
+            return
+        self._count += fillin
+        filled_len = int(round(self._bar_len * self._count / float(total)))
+        percents = round(100.0 * self._count / float(total), 1)
+        if percents > 100:
+            percents = 100.0
+        if filled_len > self._bar_len:
+            filled_len = self._bar_len
+            
+        bar = self._symbol_done * filled_len + \
+              self._symbol_space * (self._bar_len - filled_len)
+        sys.stdout.write(' [%s] %s%s ...%s \r' % (bar, percents, '%', self._status))
+        sys.stdout.flush()
+        self._blinker += 1
+
+    def close(self):
+        if not self._cancelled:
+            self._progress(self._total, self._total)
+
+    def cancel(self):
+        self._cancelled = True
+    
+    def is_cancelled(self):
+        return self._cancelled

pytchat/cli/progressbar_simple.py

@@ -0,0 +1,49 @@
+'''
+This code for this progress bar is based on
+vladignatyev/progress.py
+https://gist.github.com/vladignatyev/06860ec2040cb497f0f3
+(MIT License)
+'''
+import sys
+
+
+class ProgressBar:
+    def __init__(self, total, status):
+        self._bar_len = 60
+        self._cancelled = False
+        print(''.join([' ' * 10, '|', '-' * (self._bar_len), '|']), end="")
+        self.reset(total=total, status=status)
+
+    def reset(self, symbol_done="=", symbol_space=" ", total=100, status=''):
+        self._symbol_done = symbol_done
+        self._symbol_space = symbol_space
+        self._total = total
+        self._status = status
+        self._old_len = 0
+        self._count = 0
+        print()
+        print(f'{status:<11}', end='')
+
+    def _disp(self, _, fetched):
+        self._progress(fetched, self._total)
+
+    def _progress(self, fillin, total):
+        if total == 0 or self._cancelled:
+            return
+        self._count += fillin
+        filled_len = int(round(self._bar_len * self._count / float(total)))
+        if filled_len > self._bar_len:
+            filled_len = self._bar_len
+        print((filled_len - self._old_len) * self._symbol_done, end="")
+        sys.stdout.flush()
+        self._old_len = filled_len
+
+    def close(self):
+        if not self._cancelled:
+            self._progress(self._total, self._total)
+
+    def cancel(self):
+        self._cancelled = True
+    
+    def is_cancelled(self):
+        return self._cancelled

pytchat/config/__init__.py

@@ -1,7 +1,8 @@
 import logging
 from . import mylogger
 headers = {
-    'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/83.0.4103.106 Safari/537.36'}
+    'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.135 Safari/537.36',
+}
 
 
 def logger(module_name: str, loglevel=None):

pytchat/core_async/livechat.py

@@ -1,13 +1,13 @@
-import aiohttp
+
 import asyncio
+import httpx
 import json
 import signal
 import time
 import traceback
 import urllib.parse
-from aiohttp.client_exceptions import ClientConnectorError
-from concurrent.futures import CancelledError
 from asyncio import Queue
+from concurrent.futures import CancelledError
 from .buffer import Buffer
 from ..parser.live import Parser
 from .. import config
@@ -22,7 +22,7 @@ MAX_RETRY = 10
 
 
 class LiveChatAsync:
-    '''asyncio(aiohttp)を利用してYouTubeのライブ配信のチャットデータを取得する。
+    '''asyncioを利用してYouTubeのライブ配信のチャットデータを取得する。
 
     Parameter
     ---------
@@ -161,11 +161,11 @@ class LiveChatAsync:
             parameter for next chat data
         '''
         try:
-            async with aiohttp.ClientSession() as session:
+            async with httpx.AsyncClient(http2=True) as client:
                 while(continuation and self._is_alive):
                     continuation = await self._check_pause(continuation)
                     contents = await self._get_contents(
-                        continuation, session, headers)
+                        continuation, client, headers)
                     metadata, chatdata = self._parser.parse(contents)
 
                     timeout = metadata['timeoutMs'] / 1000
@@ -210,7 +210,7 @@ class LiveChatAsync:
                     self._video_id, 3, self._topchat_only)
         return continuation
 
-    async def _get_contents(self, continuation, session, headers):
+    async def _get_contents(self, continuation, client, headers):
         '''Get 'continuationContents' from livechat json.
            If contents is None at first fetching,
            try to fetch archive chat data.
@@ -219,7 +219,7 @@ class LiveChatAsync:
           -------
             'continuationContents' which includes metadata & chatdata.
         '''
-        livechat_json = await self._get_livechat_json(continuation, session, headers)
+        livechat_json = await self._get_livechat_json(continuation, client, headers)
         contents = self._parser.get_contents(livechat_json)
         if self._first_fetch:
             if contents is None or self._is_replay:
@@ -229,18 +229,18 @@ class LiveChatAsync:
                 continuation = arcparam.getparam(
                     self._video_id, self.seektime, self._topchat_only)
                 livechat_json = (await self._get_livechat_json(
-                                 continuation, session, headers))
+                                 continuation, client, headers))
                 reload_continuation = self._parser.reload_continuation(
                     self._parser.get_contents(livechat_json))
                 if reload_continuation:
                     livechat_json = (await self._get_livechat_json(
-                        reload_continuation, session, headers))
+                        reload_continuation, client, headers))
                 contents = self._parser.get_contents(livechat_json)
                 self._is_replay = True
             self._first_fetch = False
         return contents
 
-    async def _get_livechat_json(self, continuation, session, headers):
+    async def _get_livechat_json(self, continuation, client, headers):
         '''
         Get json which includes chat data.
         '''
@@ -249,14 +249,13 @@ class LiveChatAsync:
         status_code = 0
         url = f"https://www.youtube.com/{self._fetch_url}{continuation}&pbj=1"
         for _ in range(MAX_RETRY + 1):
-            async with session.get(url, headers=headers) as resp:
+            try:
-                try:
+                resp = await client.get(url, headers=headers)
-                    text = await resp.text()
+                livechat_json = resp.json()
-                    livechat_json = json.loads(text)
+                break
-                    break
+            except (httpx.HTTPError, json.JSONDecodeError):
-                except (ClientConnectorError, json.JSONDecodeError):
+                await asyncio.sleep(1)
-                    await asyncio.sleep(1)
+                continue
-                    continue
         else:
             self._logger.error(f"[{self._video_id}]"
                                f"Exceeded retry count. status_code={status_code}")

pytchat/core_multithread/livechat.py

@@ -1,4 +1,4 @@
-import requests
+import httpx
 import json
 import signal
 import time
@@ -153,10 +153,10 @@ class LiveChat:
             parameter for next chat data
         '''
         try:
-            with requests.Session() as session:
+            with httpx.Client(http2=True) as client:
                 while(continuation and self._is_alive):
                     continuation = self._check_pause(continuation)
-                    contents = self._get_contents(continuation, session, headers)
+                    contents = self._get_contents(continuation, client, headers)
                     metadata, chatdata = self._parser.parse(contents)
                     timeout = metadata['timeoutMs'] / 1000
                     chat_component = {
@@ -199,7 +199,7 @@ class LiveChat:
                 continuation = liveparam.getparam(self._video_id, 3)
         return continuation
 
-    def _get_contents(self, continuation, session, headers):
+    def _get_contents(self, continuation, client, headers):
         '''Get 'continuationContents' from livechat json.
            If contents is None at first fetching,
            try to fetch archive chat data.
@@ -209,7 +209,7 @@ class LiveChat:
             'continuationContents' which includes metadata & chat data.
         '''
         livechat_json = (
-            self._get_livechat_json(continuation, session, headers)
+            self._get_livechat_json(continuation, client, headers)
         )
         contents = self._parser.get_contents(livechat_json)
         if self._first_fetch:
@@ -219,18 +219,18 @@ class LiveChat:
                 self._fetch_url = "live_chat_replay/get_live_chat_replay?continuation="
                 continuation = arcparam.getparam(
                     self._video_id, self.seektime, self._topchat_only)
-                livechat_json = (self._get_livechat_json(continuation, session, headers))
+                livechat_json = (self._get_livechat_json(continuation, client, headers))
                 reload_continuation = self._parser.reload_continuation(
                     self._parser.get_contents(livechat_json))
                 if reload_continuation:
                     livechat_json = (self._get_livechat_json(
-                        reload_continuation, session, headers))
+                        reload_continuation, client, headers))
                 contents = self._parser.get_contents(livechat_json)
                 self._is_replay = True
             self._first_fetch = False
         return contents
 
-    def _get_livechat_json(self, continuation, session, headers):
+    def _get_livechat_json(self, continuation, client, headers):
         '''
         Get json which includes chat data.
         '''
@@ -239,10 +239,9 @@ class LiveChat:
         status_code = 0
         url = f"https://www.youtube.com/{self._fetch_url}{continuation}&pbj=1"
         for _ in range(MAX_RETRY + 1):
-            with session.get(url, headers=headers) as resp:
+            with client:
                 try:
-                    text = resp.text
+                    livechat_json = client.get(url, headers=headers).json()
-                    livechat_json = json.loads(text)
                     break
                 except json.JSONDecodeError:
                     time.sleep(1)

pytchat/exceptions.py

@@ -62,3 +62,18 @@ class ReceivedUnknownContinuation(ChatParseException):
 
 class FailedExtractContinuation(ChatDataFinished):
     pass
+
+
+class VideoInfoParseError(Exception):
+    '''
+    thrown when failed to parse video info
+    '''
+
+
+class PatternUnmatchError(VideoInfoParseError):
+    '''
+    thrown when failed to parse video info with unmatched pattern
+    '''
+    def __init__(self, doc):
+        self.msg = "PatternUnmatchError"
+        self.doc = doc

pytchat/processors/default/renderer/paidsticker.py

@@ -21,7 +21,7 @@ class LiveChatPaidStickerRenderer(BaseRenderer):
         self.amountString = amountDisplayString
         self.currency = currency.symbols[symbol]["fxtext"] if currency.symbols.get(
             symbol) else symbol
-        self.bgColor = self.renderer.get("moneyChipBackgroundColor", 0)
+        self.bgColor = self.renderer.get("backgroundColor", 0)
         self.sticker = "".join(("https:",
             self.renderer["sticker"]["thumbnails"][0]["url"]))
         self.colors = self.get_colors()

pytchat/processors/html_archiver.py

@@ -1,6 +1,6 @@
 import os
 import re
-import requests
+import httpx
 from base64 import standard_b64encode
 from .chat_processor import ChatProcessor
 from .default.processor import DefaultProcessor
@@ -43,20 +43,21 @@ class HTMLArchiver(ChatProcessor):
     '''
     HTMLArchiver saves chat data as HTML table format.
     '''
-    def __init__(self, save_path):
+    def __init__(self, save_path, callback=None):
         super().__init__()
         self.save_path = self._checkpath(save_path)
         self.processor = DefaultProcessor()
         self.emoji_table = {}  # tuble for custom emojis. key: emoji_id, value: base64 encoded image binary.
         self.header = [HEADER_HTML]
         self.body = ['<body>\n', '<table class="css">\n', self._parse_table_header(fmt_headers)]
+        self.callback = callback
 
     def _checkpath(self, filepath):
         splitter = os.path.splitext(os.path.basename(filepath))
         body = splitter[0]
         extention = splitter[1]
         newpath = filepath
-        counter = 0
+        counter = 1
         while os.path.exists(newpath):
             match = re.search(PATTERN, body)
             if match:
@@ -80,17 +81,20 @@ class HTMLArchiver(ChatProcessor):
         """
         if chat_components is None or len(chat_components) == 0:
             return
-        self.body.extend(
+        for c in self.processor.process(chat_components).items:
-            (self._parse_html_line((
+            self.body.extend(
-                c.datetime,
+                self._parse_html_line((
-                c.elapsedTime,
+                    c.datetime,
-                c.author.name,
+                    c.elapsedTime,
-                self._parse_message(c.messageEx),
+                    c.author.name,
-                c.amountString,
+                    self._parse_message(c.messageEx),
-                c.author.type,
+                    c.amountString,
-                c.author.channelId)
+                    c.author.type,
-            ) for c in self.processor.process(chat_components).items)
+                    c.author.channelId)
-        )
+                )
+            )
+            if self.callback:
+                self.callback(None, 1)
 
     def _parse_html_line(self, raw_line):
         return ''.join(('<tr>',
@@ -108,7 +112,7 @@ class HTMLArchiver(ChatProcessor):
                        for item in message_items)
 
     def _encode_img(self, url):
-        resp = requests.get(url)
+        resp = httpx.get(url)
         return standard_b64encode(resp.content).decode()
 
     def _set_emoji_table(self, item: dict):
@@ -131,7 +135,7 @@ class HTMLArchiver(ChatProcessor):
     
     def finalize(self):
         self.header.extend([self._create_styles(), '</head>\n'])
-        self.body.extend(['</table>\n</body>'])
+        self.body.extend(['</table>\n</body>\n</html>'])
         with open(self.save_path, mode='a', encoding='utf-8') as f:
             f.writelines(self.header)
             f.writelines(self.body)

pytchat/tool/extract/asyncdl.py

@@ -1,6 +1,5 @@
-import aiohttp
+import httpx
 import asyncio
-import json
 from . import parser
 from . block import Block
 from . worker import ExtractWorker
@@ -55,7 +54,7 @@ def ready_blocks(video_id, duration, div, callback):
         raise ValueError
 
     async def _get_blocks(video_id, duration, div, callback):
-        async with aiohttp.ClientSession() as session:
+        async with httpx.AsyncClient(http2=True) as session:
             tasks = [_create_block(session, video_id, seektime, callback)
                      for seektime in _split(-1, duration, div)]
             return await asyncio.gather(*tasks)
@@ -65,9 +64,8 @@ def ready_blocks(video_id, duration, div, callback):
         url = f"{REPLAY_URL}{quote(continuation)}&pbj=1"
         for _ in range(MAX_RETRY_COUNT):
             try:
-                async with session.get(url, headers=headers) as resp:
+                resp = await session.get(url, headers=headers)
-                    text = await resp.text()
+                next_continuation, actions = parser.parse(resp.json())
-                next_continuation, actions = parser.parse(json.loads(text))
                 break
             except JSONDecodeError:
                 await asyncio.sleep(3)
@@ -106,7 +104,7 @@ def fetch_patch(callback, blocks, video_id):
             )
             for block in blocks
         ]
-        async with aiohttp.ClientSession() as session:
+        async with httpx.AsyncClient() as session:
             tasks = [worker.run(session) for worker in workers]
             return await asyncio.gather(*tasks)
 
@@ -114,9 +112,8 @@ def fetch_patch(callback, blocks, video_id):
         url = f"{REPLAY_URL}{quote(continuation)}&pbj=1"
         for _ in range(MAX_RETRY_COUNT):
             try:
-                async with session.get(url, headers=config.headers) as resp:
+                resp = await session.get(url, headers=config.headers)
-                    chat_json = await resp.text()
+                continuation, actions = parser.parse(resp.json())
-                continuation, actions = parser.parse(json.loads(chat_json))
                 break
             except JSONDecodeError:
                 await asyncio.sleep(3)

pytchat/tool/extract/extractor.py

@@ -79,7 +79,7 @@ class Extractor:
 
     def extract(self):
         if self.duration == 0:
-            print("video is not archived.")
+            print("\nCannot extract chat data:\n The specified video has not yet been archived.")
             return []
         data = self._execute_extract_operations()
         if self.processor is None:

pytchat/tool/extract/parser.py

@@ -42,10 +42,14 @@ def get_offset(item):
 
 
 def get_id(item):
-    return list((list(item['replayChatItemAction']["actions"][0].values()
+    a = list(item['replayChatItemAction']["actions"][0].values())[0].get('item')
-                      )[0])['item'].values())[0].get('id')
+    if a:
+        return list(a.values())[0].get('id')
+    return None
 
 
 def get_type(item):
-    return list((list(item['replayChatItemAction']["actions"][0].values()
+    a = list(item['replayChatItemAction']["actions"][0].values())[0].get('item')
-                      )[0])['item'].keys())[0]
+    if a:
+        return list(a.keys())[0]
+    return None

pytchat/tool/extract/worker.py

@@ -1,6 +1,7 @@
 from . block import Block
 from . patch import fill, split
 from ... paramgen import arcparam
+from typing import Tuple
 
 
 class ExtractWorker:
@@ -76,7 +77,7 @@ def _search_new_block(worker) -> Block:
     return new_block
 
 
-def _get_undone_block(blocks) -> (int, Block):
+def _get_undone_block(blocks) -> Tuple[int, Block]:
     min_interval_ms = 120000
     max_remaining = 0
     undone_block = None

pytchat/tool/mining/asyncdl.py

@@ -1,141 +0,0 @@
-
-import aiohttp
-import asyncio
-import json
-from . import parser
-from . block import Block
-from . worker import ExtractWorker
-from . patch import Patch
-from ... import config 
-from ... paramgen import arcparam_mining as arcparam
-from concurrent.futures import CancelledError
-from urllib.parse import quote
-
-headers = config.headers
-REPLAY_URL = "https://www.youtube.com/live_chat_replay?continuation="
-INTERVAL = 1
-def _split(start, end, count, min_interval_sec = 120):
-    """
-    Split section from `start` to `end` into `count` pieces,
-    and returns the beginning of each piece. 
-    The `count` is adjusted so that the length of each piece
-    is no smaller than `min_interval`.
-
-    Returns:
-    --------
-        List of the offset of each block's first chat data.
-    """
-    
-    if not (isinstance(start,int) or isinstance(start,float)) or \
-       not (isinstance(end,int) or isinstance(end,float)):
-        raise ValueError("start/end must be int or float")
-    if not isinstance(count,int):
-        raise ValueError("count must be int")
-    if start>end:
-        raise ValueError("end must be equal to or greater than start.")
-    if count<1:
-        raise ValueError("count must be equal to or greater than 1.")
-    if (end-start)/count < min_interval_sec:
-        count = int((end-start)/min_interval_sec) 
-        if count == 0 : count = 1
-    interval= (end-start)/count 
-    
-    if count == 1:
-        return [start]
-    return sorted( list(set( [int(start + interval*j)
-        for j in range(count) ])))
-
-def ready_blocks(video_id, duration, div, callback):
-    if div <= 0: raise ValueError
-
-    async def _get_blocks( video_id, duration, div, callback):
-        async with aiohttp.ClientSession() as session:
-            tasks = [_create_block(session, video_id,  seektime, callback)
-                for  seektime in _split(0, duration, div)]
-            return await asyncio.gather(*tasks)
-
-    
-
-    async def _create_block(session, video_id, seektime, callback):
-        continuation = arcparam.getparam(video_id, seektime = seektime)
-        url=(f"{REPLAY_URL}{quote(continuation)}&playerOffsetMs="
-            f"{int(seektime*1000)}&hidden=false&pbj=1")
-        async with session.get(url, headers = headers) as resp:
-            chat_json = await resp.text()
-        if chat_json is None:
-            return
-        continuation, actions = parser.parse(json.loads(chat_json)[1])
-        first = seektime
-        seektime += INTERVAL
-        if callback:
-            callback(actions, INTERVAL)
-        return Block(
-            continuation = continuation,
-            chat_data = actions,
-            first = first,
-            last = seektime,
-            seektime = seektime
-        )
-    """
-    fetch initial blocks.
-    """  
-    loop = asyncio.get_event_loop()
-    blocks = loop.run_until_complete(
-        _get_blocks(video_id, duration, div, callback))
-    return blocks
-
-def fetch_patch(callback, blocks, video_id):
-
-    async def _allocate_workers():
-        workers = [
-            ExtractWorker(
-                fetch = _fetch,  block = block,
-                blocks = blocks, video_id = video_id
-            )
-            for block in blocks
-        ]
-        async with aiohttp.ClientSession() as session:
-            tasks = [worker.run(session) for worker in workers]
-            return await asyncio.gather(*tasks)    
-
-    async def _fetch(seektime,session) -> Patch:
-        continuation = arcparam.getparam(video_id, seektime = seektime)
-        url=(f"{REPLAY_URL}{quote(continuation)}&playerOffsetMs="
-            f"{int(seektime*1000)}&hidden=false&pbj=1")
-        async with session.get(url,headers = config.headers) as resp:
-            chat_json = await resp.text()
-        actions = []
-        try:
-            if chat_json is None:
-                return Patch()
-            continuation, actions = parser.parse(json.loads(chat_json)[1])
-        except json.JSONDecodeError:
-            pass
-        if callback:
-            callback(actions, INTERVAL)
-        return Patch(chats = actions, continuation = continuation, 
-            seektime = seektime, last = seektime)
-    """
-    allocate workers and assign blocks.
-    """   
-    loop = asyncio.get_event_loop()
-    try:
-        loop.run_until_complete(_allocate_workers())
-    except CancelledError:
-        pass
-
-async def _shutdown():
-    print("\nshutdown...")
-    tasks = [t for t in asyncio.all_tasks()
-         if t is not asyncio.current_task()]
-    for task in tasks:
-        task.cancel()
-        try:
-            await task
-        except asyncio.CancelledError:
-            pass
-
-def cancel():
-    loop = asyncio.get_event_loop()
-    loop.create_task(_shutdown())
-    

pytchat/tool/mining/block.py

@@ -1,62 +0,0 @@
-from . import parser
-class Block:
-    """Block object represents something like a box 
-    to join chunk of chatdata.
-
-    Parameter:
-    ---------
-    first : int :
-        videoOffsetTimeMs of the first chat_data 
-        (chat_data[0])
-        
-    last : int :
-        videoOffsetTimeMs of the last chat_data.
-        (chat_data[-1])
-
-        this value increases as fetching chatdata progresses.
-
-    end : int :
-        target videoOffsetTimeMs of last chat data for extract,
-        equals to first videoOffsetTimeMs of next block.
-        when extract worker reaches this offset, stop fetching.
-
-    continuation : str :
-        continuation param of last chat data.
-
-    chat_data : list 
-
-    done : bool :
-        whether this block has been fetched.
-    
-    remaining : int :
-        remaining data to extract.
-        equals end - last.
-    
-    is_last : bool :
-        whether this block is the last one in blocklist.
-
-    during_split : bool :
-        whether this block is in the process of during_split.
-        while True, this block is excluded from duplicate split procedure.
-    
-    seektime : float :
-        the last position of this block(seconds) already fetched.
-    """
-    
-    __slots__ = ['first','last','end','continuation','chat_data','remaining',
-        'done','is_last','during_split','seektime']
-
-    def __init__(self, first = 0, last = 0, end = 0,
-                continuation = '', chat_data = [], is_last = False,
-                during_split = False, seektime = None):
-        self.first = first
-        self.last = last
-        self.end = end
-        self.continuation = continuation
-        self.chat_data = chat_data
-        self.done = False
-        self.remaining = self.end - self.last
-        self.is_last = is_last
-        self.during_split = during_split
-        self.seektime = seektime
- 

pytchat/tool/mining/parser.py

@@ -1,73 +0,0 @@
-import re
-from ... import config
-from ... exceptions import (
-    ResponseContextError,
-    NoContents, NoContinuation)
-
-logger = config.logger(__name__)
-
-
-def parse(jsn):
-    """
-    Parse replay chat data.
-    Parameter:
-    ----------
-    jsn : dict
-        JSON of replay chat data.
-    Returns:
-    ------
-        continuation : str
-        actions : list
-
-    """
-    if jsn is None:
-        raise ValueError("parameter JSON is None")
-    if jsn['response']['responseContext'].get('errors'):
-        raise ResponseContextError(
-            'video_id is invalid or private/deleted.')
-    contents = jsn["response"].get('continuationContents')
-    if contents is None:
-        raise NoContents('No chat data.')
-
-    cont = contents['liveChatContinuation']['continuations'][0]
-    if cont is None:
-        raise NoContinuation('No Continuation')
-    metadata = cont.get('liveChatReplayContinuationData')
-    if metadata:
-        continuation = metadata.get("continuation")
-        actions = contents['liveChatContinuation'].get('actions')
-        if continuation:
-            return continuation, [action["replayChatItemAction"]["actions"][0]
-                                  for action in actions
-                                  if list(action['replayChatItemAction']["actions"][0].values()
-                                          )[0]['item'].get("liveChatPaidMessageRenderer")
-                                  or list(action['replayChatItemAction']["actions"][0].values()
-                                          )[0]['item'].get("liveChatPaidStickerRenderer")
-                                  ]
-    return None, []
-
-
-def get_offset(item):
-    return int(item['replayChatItemAction']["videoOffsetTimeMsec"])
-
-
-def get_id(item):
-    return list((list(item['replayChatItemAction']["actions"][0].values()
-                      )[0])['item'].values())[0].get('id')
-
-
-def get_type(item):
-    return list((list(item['replayChatItemAction']["actions"][0].values()
-                      )[0])['item'].keys())[0]
-
-
-_REGEX_YTINIT = re.compile(
-    "window\\[\"ytInitialData\"\\]\\s*=\\s*({.+?});\\s+")
-
-
-def extract(text):
-
-    match = re.findall(_REGEX_YTINIT, str(text))
-    if match:
-        return match[0]
-    return None

pytchat/tool/mining/patch.py

@@ -1,27 +0,0 @@
-from . import parser
-from . block import Block
-from typing import NamedTuple
-
-class Patch(NamedTuple):
-    """
-    Patch represents chunk of chat data
-    which is fetched by asyncdl.fetch_patch._fetch().
-    """
-    chats : list = []
-    continuation : str = None
-    seektime : float = None
-    first : int = None
-    last : int = None
-
-def fill(block:Block, patch:Patch):
-    if patch.last < block.end:
-        set_patch(block, patch)
-        return
-    block.continuation = None
-
-def set_patch(block:Block, patch:Patch):
-    block.continuation = patch.continuation
-    block.chat_data.extend(patch.chats)
-    block.last = patch.seektime
-    block.seektime = patch.seektime       
-

pytchat/tool/mining/superchat_miner.py

@@ -1,72 +0,0 @@
-from . import asyncdl
-from . import parser
-from .. videoinfo import VideoInfo
-from ... import config
-from ... exceptions import InvalidVideoIdException
-logger = config.logger(__name__)
-headers=config.headers
-
-class SuperChatMiner:
-    def __init__(self, video_id, duration, div, callback):
-        if not isinstance(div ,int) or div < 1:
-            raise ValueError('div must be positive integer.')
-        elif div > 10:
-            div = 10
-        if not isinstance(duration ,int) or duration < 1:
-            raise ValueError('duration must be positive integer.')
-        self.video_id = video_id
-        self.duration = duration
-        self.div = div
-        self.callback = callback
-        self.blocks = []
-
-    def _ready_blocks(self):
-        blocks = asyncdl.ready_blocks(
-            self.video_id, self.duration, self.div, self.callback)
-        self.blocks = [block for block in blocks if block is not None]
-        return self  
-
-    def _set_block_end(self):
-        for i in range(len(self.blocks)-1):
-            self.blocks[i].end = self.blocks[i+1].first
-        self.blocks[-1].end = self.duration
-        self.blocks[-1].is_last =True
-        return self
-
-    def _download_blocks(self):
-        asyncdl.fetch_patch(self.callback, self.blocks, self.video_id)
-        return self
-
-    def _combine(self):
-        ret = []
-        for block in self.blocks:
-            ret.extend(block.chat_data) 
-        return ret
-
-    def extract(self):
-        return (
-            self._ready_blocks()
-                ._set_block_end()
-                ._download_blocks()
-                ._combine()
-        )
-
-def extract(video_id, div = 1, callback = None, processor = None):
-    duration = 0
-    try:
-        duration = VideoInfo(video_id).get_duration()
-    except InvalidVideoIdException:
-        raise
-    if duration == 0:
-        print("video is live.")
-        return []
-    data = SuperChatMiner(video_id, duration, div, callback).extract()
-    if processor is None:
-        return data
-    return processor.process(
-        [{'video_id':None,'timeout':1,'chatdata' : (action
-        for action in data)}]
-    )
-
-def cancel():
-    asyncdl.cancel()

pytchat/tool/mining/worker.py

@@ -1,45 +0,0 @@
-from . import parser
-from . block import Block
-from . patch import Patch, fill
-from ... paramgen import arcparam
-INTERVAL = 1
-class ExtractWorker:
-    """
-    ExtractWorker associates a download session with a block.
-
-    When the worker finishes fetching, the block
-    being fetched is splitted and assigned the free worker.
-
-    Parameter
-    ----------
-    fetch : func :
-        extract function of asyncdl
-
-    block : Block :
-        Block object that includes chat_data
-    
-    blocks : list :
-        List of Block(s)
-
-    video_id : str :
-
-    parent_block : Block :
-        the block from which current block is splitted 
-    """
-    __slots__ = ['block', 'fetch', 'blocks', 'video_id', 'parent_block']
-    def __init__(self, fetch, block, blocks, video_id ):
-        self.block:Block = block
-        self.fetch = fetch
-        self.blocks:list = blocks
-        self.video_id:str = video_id
-        self.parent_block:Block = None
-
-    async def run(self, session):
-        while self.block.continuation:
-            patch = await self.fetch(
-                self.block.seektime, session)
-            fill(self.block, patch)
-            self.block.seektime += INTERVAL
-        self.block.done = True    
-
-

pytchat/tool/videoinfo.py

@@ -1,13 +1,13 @@
 import json
 import re
-import requests
+import httpx
 from .. import config
-from ..exceptions import InvalidVideoIdException
+from ..exceptions import InvalidVideoIdException, PatternUnmatchError
 from ..util.extract_video_id import extract_video_id
 
 headers = config.headers
 
-pattern = re.compile(r"yt\.setConfig\({'PLAYER_CONFIG': ({.*})}\);")
+pattern = re.compile(r"'PLAYER_CONFIG': ({.*}}})")
 
 item_channel_id = [
     "videoDetails",
@@ -85,13 +85,16 @@ class VideoInfo:
 
     def _get_page_text(self, video_id):
         url = f"https://www.youtube.com/embed/{video_id}"
-        resp = requests.get(url, headers=headers)
+        resp = httpx.get(url, headers=headers)
         resp.raise_for_status()
         return resp.text
 
     def _parse(self, text):
         result = re.search(pattern, text)
-        res = json.loads(result.group(1))
+        if result is None:
+            raise PatternUnmatchError(text)
+        decoder = json.JSONDecoder()
+        res = decoder.raw_decode(result.group(1)[:-1])[0]
         response = self._get_item(res, item_response)
         if response is None:
             self._check_video_is_private(res.get("args"))

pytchat/util/__init__.py

@@ -1,11 +1,11 @@
-import requests
+import httpx
 import json
 import datetime
 from .. import config
 
 
 def extract(url):
-    _session = requests.Session()
+    _session = httpx.Client(http2=True)
     html = _session.get(url, headers=config.headers)
     with open(str(datetime.datetime.now().strftime('%Y-%m-%d %H-%M-%S')
                   ) + 'test.json', mode='w', encoding='utf-8') as f:

requirements.txt

@@ -1,5 +1,4 @@
-aiohttp
+httpx[http2]==0.14.1
-protobuf
+protobuf==3.13.0
 pytz
-requests
 urllib3

requirements_test.txt

@@ -1,5 +1,4 @@
-aioresponses
 mock
 mocker
 pytest
-pytest-mock
+pytest_httpx

tests/test_arcparam.py

@@ -1,5 +1,5 @@
 import json
-import requests
+import httpx
 import pytchat.config as config
 from pytchat.paramgen import arcparam
 from pytchat.parser.live import Parser
@@ -18,14 +18,15 @@ def test_arcparam_1(mocker):
 def test_arcparam_2(mocker):
     param = arcparam.getparam("SsjCnHOk-Sk", seektime=100)
     url = f"https://www.youtube.com/live_chat_replay/get_live_chat_replay?continuation={param}&pbj=1"
-    resp = requests.Session().get(url, headers=config.headers)
+    resp = httpx.Client(http2=True).get(url, headers=config.headers)
     jsn = json.loads(resp.text)
     parser = Parser(is_replay=True)
     contents = parser.get_contents(jsn)
-    _ , chatdata = parser.parse(contents)
+    _, chatdata = parser.parse(contents)
     test_id = chatdata[0]["addChatItemAction"]["item"]["liveChatTextMessageRenderer"]["id"]
     assert test_id == "CjoKGkNMYXBzZTdudHVVQ0Zjc0IxZ0FkTnFnQjVREhxDSnlBNHV2bnR1VUNGV0dnd2dvZDd3NE5aZy0w"
 
+
 def test_arcparam_3(mocker):
     param = arcparam.getparam("01234567890")
     assert param == "op2w0wQmGhxDZzhLRFFvTE1ERXlNelExTmpjNE9UQWdBUT09SARgAXICCAE%3D"

tests/test_arcparam_mining.py

@@ -1,41 +0,0 @@
-from pytchat.tool.mining import parser
-import pytchat.config as config
-import requests
-import json
-from pytchat.paramgen import arcparam_mining as arcparam
-
-
-def test_arcparam_e(mocker):
-    try:
-        arcparam.getparam("01234567890", -1)
-        assert False
-    except ValueError:
-        assert True
-
-
-def test_arcparam_0(mocker):
-    param = arcparam.getparam("01234567890", 0)
-
-    assert param == "op2w0wQsGiBDZzhhRFFvTE1ERXlNelExTmpjNE9UQWdBUSUzRCUzREABYARyAggBeAE%3D"
-
-
-def test_arcparam_1(mocker):
-    param = arcparam.getparam("01234567890", seektime=100000)
-    print(param)
-    assert param == "op2w0wQzGiBDZzhhRFFvTE1ERXlNelExTmpjNE9UQWdBUSUzRCUzREABWgUQgMLXL2AEcgIIAXgB"
-
-
-def test_arcparam_2(mocker):
-    param = arcparam.getparam("PZz9NB0-Z64", 1)
-    url = f"https://www.youtube.com/live_chat_replay?continuation={param}&playerOffsetMs=1000&pbj=1"
-    resp = requests.Session().get(url, headers=config.headers)
-    jsn = json.loads(resp.text)
-    _, chatdata = parser.parse(jsn[1])
-    test_id = chatdata[0]["addChatItemAction"]["item"]["liveChatPaidMessageRenderer"]["id"]
-    print(test_id)
-    assert test_id == "ChwKGkNKSGE0YnFJeWVBQ0ZWcUF3Z0VkdGIwRm9R"
-
-
-def test_arcparam_3(mocker):
-    param = arcparam.getparam("01234567890")
-    assert param == "op2w0wQsGiBDZzhhRFFvTE1ERXlNelExTmpjNE9UQWdBUSUzRCUzREABYARyAggBeAE%3D"

tests/test_default_processor.py

@@ -136,7 +136,7 @@ def test_supersticker(mocker):
     assert ret.amountValue == 200
     assert ret.amountString == "￥200"
     assert ret.currency == "JPY"
-    assert ret.bgColor == 4278248959
+    assert ret.bgColor == 4278237396
     assert ret.sticker == "https://lh3.googleusercontent.com/param_s=s72-rp"
     assert ret.author.name == "author_name"
     assert ret.author.channelId == "author_channel_id"

tests/test_extract_asyncdl.py

@@ -1,77 +0,0 @@
-import aiohttp
-import asyncio
-import json
-from pytchat.tool.extract import parser
-import sys
-import time
-from aioresponses import aioresponses
-from concurrent.futures import CancelledError
-from pytchat.tool.extract import asyncdl
-
-def _open_file(path):
-    with open(path,mode ='r',encoding = 'utf-8') as f:
-        return f.read()
-
-
-def test_asyncdl_split():
-
-    ret = asyncdl._split(0,1000,1)
-    assert ret == [0]
-
-    ret = asyncdl._split(1000,1000,10)
-    assert ret == [1000]
-
-    ret = asyncdl._split(0,1000,5)
-    assert ret == [0,200,400,600,800]
-
-    ret = asyncdl._split(10.5, 700.3, 5)
-    assert ret == [10, 148, 286, 424, 562]
-
-
-    ret = asyncdl._split(0,500,5)
-    assert ret == [0,125,250,375]
-    
-    ret = asyncdl._split(0,500,500)
-    assert ret == [0,125,250,375]
-    
-    ret = asyncdl._split(-1,1000,5)
-    assert ret == [-1, 199, 399, 599, 799]
-    
-    """invalid argument order"""
-    try:
-        ret = asyncdl._split(500,0,5)
-        assert False
-    except ValueError:
-        assert True
-
-    """invalid count"""
-    try:
-        ret = asyncdl._split(0,500,-1)
-        assert False
-    except ValueError:
-        assert True
-
-    try:
-        ret = asyncdl._split(0,500,0)
-        assert False
-    except ValueError:
-        assert True
-
-    """invalid argument type"""
-    try:
-        ret = asyncdl._split(0,5000,5.2)
-        assert False
-    except ValueError:
-        assert True
-
-    try:
-        ret = asyncdl._split(0,5000,"test")
-        assert False
-    except ValueError:
-        assert True
-
-    try:
-        ret = asyncdl._split([0,1],5000,5)
-        assert False
-    except ValueError:
-        assert True

tests/test_extract_duplcheck.py

@@ -1,60 +1,66 @@
-import aiohttp
-import asyncio
 import json
-import os, sys
-import time
 from pytchat.tool.extract import duplcheck
 from pytchat.tool.extract import parser
 from pytchat.tool.extract.block import Block
 from pytchat.tool.extract.duplcheck import _dump
-def _open_file(path):
-    with open(path,mode ='r',encoding = 'utf-8') as f:
-        return f.read()
 
 
+def _open_file(path):
+    with open(path, mode='r', encoding='utf-8') as f:
+        return f.read()
+
 
 def test_overlap():
     """
-    test overlap data 
+    test overlap data
         operation : [0]  [2] [3]  [4] -> last :align to end
                     [1] , [5] -> no change
-        
+
     """
 
     def load_chatdata(filename):
         return parser.parse(
-            json.loads(_open_file("tests/testdata/extract_duplcheck/overlap/"+filename))
+            json.loads(_open_file(
+                "tests/testdata/extract_duplcheck/overlap/" + filename))
         )[1]
 
     blocks = (
-        Block(first = 0,    last= 12771, end=  9890,chat_data = load_chatdata("dp0-0.json")),     
+        Block(first=0, last=12771, end=9890,
-        Block(first = 9890, last= 15800, end= 20244,chat_data = load_chatdata("dp0-1.json")), 
+              chat_data=load_chatdata("dp0-0.json")),
-        Block(first = 20244,last= 45146, end= 32476,chat_data = load_chatdata("dp0-2.json")), 
+        Block(first=9890, last=15800, end=20244,
-        Block(first = 32476,last= 50520, end= 41380,chat_data = load_chatdata("dp0-3.json")), 
+              chat_data=load_chatdata("dp0-1.json")),
-        Block(first = 41380,last= 62875, end= 52568,chat_data = load_chatdata("dp0-4.json")), 
+        Block(first=20244, last=45146, end=32476,
-        Block(first = 52568,last= 62875, end= 54000,chat_data = load_chatdata("dp0-5.json"),is_last=True)
+              chat_data=load_chatdata("dp0-2.json")),
+        Block(first=32476, last=50520, end=41380,
+              chat_data=load_chatdata("dp0-3.json")),
+        Block(first=41380, last=62875, end=52568,
+              chat_data=load_chatdata("dp0-4.json")),
+        Block(first=52568, last=62875, end=54000,
+              chat_data=load_chatdata("dp0-5.json"), is_last=True)
     )
     result = duplcheck.remove_overlap(blocks)
-    #dp0-0.json has item offset time is 9890 (equals block[0].end = block[1].first), 
+    # dp0-0.json has item offset time is 9890 (equals block[0].end = block[1].first),
-    #but must be aligne to the most close and smaller value:9779.
+    # but must be aligne to the most close and smaller value:9779.
     assert result[0].last == 9779
-    
+
     assert result[1].last == 15800
-    
+
     assert result[2].last == 32196
-    
+
     assert result[3].last == 41116
-    
+
     assert result[4].last == 52384
-    
+
-    #the last block must be always added to result.
+    # the last block must be always added to result.
     assert result[5].last == 62875
-    
+
+
 def test_duplicate_head():
 
     def load_chatdata(filename):
         return parser.parse(
-            json.loads(_open_file("tests/testdata/extract_duplcheck/head/"+filename))
+            json.loads(_open_file(
+                "tests/testdata/extract_duplcheck/head/" + filename))
         )[1]
 
     """
@@ -69,25 +75,26 @@ def test_duplicate_head():
         result    : [2] , [4] , [5] 
     """
 
-    #chat data offsets are ignored.
+    # chat data offsets are ignored.
     blocks = (
-        Block(first = 0, last = 2500, chat_data = load_chatdata("dp0-0.json")),     
+        Block(first=0, last=2500, chat_data=load_chatdata("dp0-0.json")),
-        Block(first = 0, last =38771, chat_data = load_chatdata("dp0-1.json")), 
+        Block(first=0, last=38771, chat_data=load_chatdata("dp0-1.json")),
-        Block(first = 0, last =45146, chat_data = load_chatdata("dp0-2.json")), 
+        Block(first=0, last=45146, chat_data=load_chatdata("dp0-2.json")),
-        Block(first = 20244, last =60520, chat_data = load_chatdata("dp0-3.json")), 
+        Block(first=20244, last=60520, chat_data=load_chatdata("dp0-3.json")),
-        Block(first = 20244, last =62875, chat_data = load_chatdata("dp0-4.json")), 
+        Block(first=20244, last=62875, chat_data=load_chatdata("dp0-4.json")),
-        Block(first = 52568, last =62875, chat_data = load_chatdata("dp0-5.json"))
+        Block(first=52568, last=62875, chat_data=load_chatdata("dp0-5.json"))
     )
     _dump(blocks)
     result = duplcheck.remove_duplicate_head(blocks)
-    
+
     assert len(result) == 3
     assert result[0].first == blocks[2].first
-    assert result[0].last  == blocks[2].last
+    assert result[0].last == blocks[2].last
     assert result[1].first == blocks[4].first
-    assert result[1].last  == blocks[4].last
+    assert result[1].last == blocks[4].last
     assert result[2].first == blocks[5].first
-    assert result[2].last  == blocks[5].last
+    assert result[2].last == blocks[5].last
+
 
 def test_duplicate_tail():
     """
@@ -103,26 +110,25 @@ def test_duplicate_tail():
     """
     def load_chatdata(filename):
         return parser.parse(
-            json.loads(_open_file("tests/testdata/extract_duplcheck/head/"+filename))
+            json.loads(_open_file(
+                "tests/testdata/extract_duplcheck/head/" + filename))
         )[1]
-    #chat data offsets are ignored.    
+    # chat data offsets are ignored.
     blocks = (
-        Block(first = 0,last = 2500, chat_data=load_chatdata("dp0-0.json")),     
+        Block(first=0, last=2500, chat_data=load_chatdata("dp0-0.json")),
-        Block(first = 1500,last = 2500, chat_data=load_chatdata("dp0-1.json")), 
+        Block(first=1500, last=2500, chat_data=load_chatdata("dp0-1.json")),
-        Block(first = 10000,last = 45146, chat_data=load_chatdata("dp0-2.json")), 
+        Block(first=10000, last=45146, chat_data=load_chatdata("dp0-2.json")),
-        Block(first = 20244,last = 45146, chat_data=load_chatdata("dp0-3.json")), 
+        Block(first=20244, last=45146, chat_data=load_chatdata("dp0-3.json")),
-        Block(first = 20244,last = 62875, chat_data=load_chatdata("dp0-4.json")), 
+        Block(first=20244, last=62875, chat_data=load_chatdata("dp0-4.json")),
-        Block(first = 52568,last = 62875, chat_data=load_chatdata("dp0-5.json"))
+        Block(first=52568, last=62875, chat_data=load_chatdata("dp0-5.json"))
     )
 
     result = duplcheck.remove_duplicate_tail(blocks)
     _dump(result)
     assert len(result) == 3
     assert result[0].first == blocks[0].first
-    assert result[0].last  == blocks[0].last
+    assert result[0].last == blocks[0].last
     assert result[1].first == blocks[2].first
-    assert result[1].last  == blocks[2].last
+    assert result[1].last == blocks[2].last
     assert result[2].first == blocks[4].first
-    assert result[2].last  == blocks[4].last
+    assert result[2].last == blocks[4].last
-
-

tests/test_extract_patch.py

@@ -1,23 +1,19 @@
-import aiohttp
-import asyncio
 import json
-import os, sys
+
-import time
-from aioresponses import aioresponses
-from pytchat.tool.extract import duplcheck
 from pytchat.tool.extract import parser
 from pytchat.tool.extract.block import Block
-from pytchat.tool.extract.patch import Patch, fill, split, set_patch
+from pytchat.tool.extract.patch import Patch, split
-from pytchat.tool.extract.duplcheck import _dump
+
 
 def _open_file(path):
-    with open(path,mode ='r',encoding = 'utf-8') as f:
+    with open(path, mode='r', encoding='utf-8') as f:
         return f.read()
 
+
 def load_chatdata(filename):
-        return parser.parse(
+    return parser.parse(
-            json.loads(_open_file("tests/testdata/fetch_patch/"+filename))
+        json.loads(_open_file("tests/testdata/fetch_patch/" + filename))
-        )[1]
+    )[1]
 
 
 def test_split_0():
@@ -61,20 +57,23 @@ def test_split_0():
      @fetched patch
                             |-- patch --|
     """
-    parent = Block(first=0, last=4000, end=60000, continuation='parent', during_split=True)
+    parent = Block(first=0, last=4000, end=60000,
-    child = Block(first=0, last=0, end=60000, continuation='mean', during_split=True)
+                   continuation='parent', during_split=True)
+    child = Block(first=0, last=0, end=60000,
+                  continuation='mean', during_split=True)
     patch = Patch(chats=load_chatdata('pt0-5.json'),
-        first=32500, last=34000, continuation='patch')
+                  first=32500, last=34000, continuation='patch')
-    
+
-    split(parent,child,patch)
+    split(parent, child, patch)
 
     assert child.continuation == 'patch'
     assert parent.last < child.first
     assert parent.end == child.first
     assert child.first < child.last
     assert child.last < child.end
-    assert parent.during_split == False
+    assert parent.during_split is False
-    assert child.during_split == False
+    assert child.during_split is False
+
 
 def test_split_1():
     """patch.first <= parent_block.last
@@ -119,14 +118,15 @@ def test_split_1():
     child = Block(first=0, last=0, end=60000, continuation='mean', during_split=True)
     patch = Patch(chats=load_chatdata('pt0-5.json'),
         first=32500, last=34000, continuation='patch')
-    
-    split(parent,child,patch)
 
-    assert parent.last == 33000 #no change
+    split(parent, child, patch)
-    assert parent.end == 60000 #no change
+
+    assert parent.last == 33000  # no change
+    assert parent.end == 60000  # no change
     assert child.continuation is None
-    assert parent.during_split == False
+    assert parent.during_split is False
-    assert child.during_split == True #exclude during_split sequence
+    assert child.during_split is True  # exclude during_split sequence
+
 
 def test_split_2():
     """child_block.end < patch.last:
@@ -174,7 +174,7 @@ def test_split_2():
     patch = Patch(chats=load_chatdata('pt0-5.json'),
         first=32500, last=34000, continuation='patch')
      
-    split(parent,child,patch)
+    split(parent, child, patch)
 
     assert child.continuation is None
     assert parent.last < child.first
@@ -182,8 +182,9 @@ def test_split_2():
     assert child.first < child.last
     assert child.last < child.end
     assert child.continuation is None
-    assert parent.during_split == False
+    assert parent.during_split is False
-    assert child.during_split == False
+    assert child.during_split is False
+
 
 def test_split_none():
     """patch.last <= parent_block.last
@@ -193,7 +194,7 @@ def test_split_none():
     and parent.block.last exceeds patch.first.
 
     In this case, fetched patch is all discarded,
-    and worker searches other processing block again. 
+    and worker searches other processing block again.
     
     ~~~~~~ before ~~~~~~
 
@@ -229,10 +230,10 @@ def test_split_none():
     patch = Patch(chats=load_chatdata('pt0-5.json'),
         first=32500, last=34000, continuation='patch')
     
-    split(parent,child,patch)
+    split(parent, child, patch)
 
-    assert parent.last == 40000 #no change
+    assert parent.last == 40000  # no change
-    assert parent.end == 60000 #no change
+    assert parent.end == 60000  # no change
     assert child.continuation is None
-    assert parent.during_split == False
+    assert parent.during_split is False
-    assert child.during_split == True #exclude during_split sequence
+    assert child.during_split is True  # exclude during_split sequence

tests/test_livechat.py

@@ -1,5 +1,8 @@
+import asyncio
 import json
-from aioresponses import aioresponses
+from pytest_httpx import HTTPXMock
+from concurrent.futures import CancelledError
+from pytchat.core_multithread.livechat import LiveChat
 from pytchat.core_async.livechat import LiveChatAsync
 from pytchat.exceptions import ResponseContextError
 
@@ -9,34 +12,37 @@ def _open_file(path):
         return f.read()
 
 
-@aioresponses()
+def add_response_file(httpx_mock: HTTPXMock, jsonfile_path: str):
-def test_Async(*mock):
+    testdata = json.loads(_open_file(jsonfile_path))
-    vid = '__test_id__'
+    httpx_mock.add_response(json=testdata)
-    _text = _open_file('tests/testdata/paramgen_firstread.json')
+
-    _text = json.loads(_text)
+
-    mock[0].get(
+def test_async(httpx_mock: HTTPXMock):
-        f"https://www.youtube.com/live_chat?v={vid}&is_popout=1", status=200, body=_text)
+    add_response_file(httpx_mock, 'tests/testdata/paramgen_firstread.json')
+
+    async def test_loop():
+        try:
+            chat = LiveChatAsync(video_id='__test_id__')
+            _ = await chat.get()
+            assert chat.is_alive()
+            chat.terminate()
+            assert not chat.is_alive()
+        except ResponseContextError:
+            assert False
+    loop = asyncio.get_event_loop()
     try:
-        chat = LiveChatAsync(video_id='__test_id__')
+        loop.run_until_complete(test_loop())
+    except CancelledError:
+        assert True
+
+
+def test_multithread(httpx_mock: HTTPXMock):
+    add_response_file(httpx_mock, 'tests/testdata/paramgen_firstread.json')
+    try:
+        chat = LiveChat(video_id='__test_id__')
+        _ = chat.get()
         assert chat.is_alive()
         chat.terminate()
         assert not chat.is_alive()
     except ResponseContextError:
-        assert not chat.is_alive()
+        assert False
-
-
-def test_MultiThread(mocker):
-    _text = _open_file('tests/testdata/paramgen_firstread.json')
-    _text = json.loads(_text)
-    responseMock = mocker.Mock()
-    responseMock.status_code = 200
-    responseMock.text = _text
-    mocker.patch('requests.Session.get').return_value = responseMock
-    try:
-        chat = LiveChatAsync(video_id='__test_id__')
-        assert chat.is_alive()
-        chat.terminate()
-        assert not chat.is_alive()
-    except ResponseContextError:
-        chat.terminate()
-        assert not chat.is_alive()

tests/test_livechat_2.py

@@ -1,6 +1,6 @@
 import asyncio
-import re
+import json
-from aioresponses import aioresponses
+from pytest_httpx import HTTPXMock
 from concurrent.futures import CancelledError
 from pytchat.core_multithread.livechat import LiveChat
 from pytchat.core_async.livechat import LiveChatAsync
@@ -12,18 +12,18 @@ def _open_file(path):
         return f.read()
 
 
-@aioresponses()
+def add_response_file(httpx_mock: HTTPXMock, jsonfile_path: str):
-def test_async_live_stream(*mock):
+    testdata = json.loads(_open_file(jsonfile_path))
+    httpx_mock.add_response(json=testdata)
 
-    async def test_loop(*mock):
+
-        pattern = re.compile(
+def test_async_live_stream(httpx_mock: HTTPXMock):
-            r'^https://www.youtube.com/live_chat/get_live_chat\?continuation=.*$')
+    add_response_file(httpx_mock, 'tests/testdata/test_stream.json')
-        _text = _open_file('tests/testdata/test_stream.json')
+
-        mock[0].get(pattern, status=200, body=_text)
+    async def test_loop():
         chat = LiveChatAsync(video_id='__test_id__', processor=DummyProcessor())
         chats = await chat.get()
         rawdata = chats[0]["chatdata"]
-        # assert fetching livachat data
         assert list(rawdata[0]["addChatItemAction"]["item"].keys())[
             0] == "liveChatTextMessageRenderer"
         assert list(rawdata[1]["addChatItemAction"]["item"].keys())[
@@ -41,25 +41,16 @@ def test_async_live_stream(*mock):
 
     loop = asyncio.get_event_loop()
     try:
-        loop.run_until_complete(test_loop(*mock))
+        loop.run_until_complete(test_loop())
     except CancelledError:
         assert True
 
 
-@aioresponses()
+def test_async_replay_stream(httpx_mock: HTTPXMock):
-def test_async_replay_stream(*mock):
+    add_response_file(httpx_mock, 'tests/testdata/finished_live.json')
-
+    add_response_file(httpx_mock, 'tests/testdata/chatreplay.json')
-    async def test_loop(*mock):
-        pattern_live = re.compile(
-            r'^https://www.youtube.com/live_chat/get_live_chat\?continuation=.*$')
-        pattern_replay = re.compile(
-            r'^https://www.youtube.com/live_chat_replay/get_live_chat_replay\?continuation=.*$')
-        # empty livechat -> switch to fetch replaychat
-        _text_live = _open_file('tests/testdata/finished_live.json')
-        _text_replay = _open_file('tests/testdata/chatreplay.json')
-        mock[0].get(pattern_live, status=200, body=_text_live)
-        mock[0].get(pattern_replay, status=200, body=_text_replay)
 
+    async def test_loop():
         chat = LiveChatAsync(video_id='__test_id__', processor=DummyProcessor())
         chats = await chat.get()
         rawdata = chats[0]["chatdata"]
@@ -71,27 +62,16 @@ def test_async_replay_stream(*mock):
 
     loop = asyncio.get_event_loop()
     try:
-        loop.run_until_complete(test_loop(*mock))
+        loop.run_until_complete(test_loop())
     except CancelledError:
         assert True
 
 
-@aioresponses()
+def test_async_force_replay(httpx_mock: HTTPXMock):
-def test_async_force_replay(*mock):
+    add_response_file(httpx_mock, 'tests/testdata/test_stream.json')
+    add_response_file(httpx_mock, 'tests/testdata/chatreplay.json')
 
-    async def test_loop(*mock):
+    async def test_loop():
-        pattern_live = re.compile(
-            r'^https://www.youtube.com/live_chat/get_live_chat\?continuation=.*$')
-        pattern_replay = re.compile(
-            r'^https://www.youtube.com/live_chat_replay/get_live_chat_replay\?continuation=.*$')
-        # valid live data, but force_replay = True
-        _text_live = _open_file('tests/testdata/test_stream.json')
-        # valid replay data
-        _text_replay = _open_file('tests/testdata/chatreplay.json')
-
-        mock[0].get(pattern_live, status=200, body=_text_live)
-        mock[0].get(pattern_replay, status=200, body=_text_replay)
-        # force replay
         chat = LiveChatAsync(
             video_id='__test_id__', processor=DummyProcessor(), force_replay=True)
         chats = await chat.get()
@@ -105,20 +85,13 @@ def test_async_force_replay(*mock):
 
     loop = asyncio.get_event_loop()
     try:
-        loop.run_until_complete(test_loop(*mock))
+        loop.run_until_complete(test_loop())
     except CancelledError:
         assert True
 
 
-def test_multithread_live_stream(mocker):
+def test_multithread_live_stream(httpx_mock: HTTPXMock):
-
+    add_response_file(httpx_mock, 'tests/testdata/test_stream.json')
-    _text = _open_file('tests/testdata/test_stream.json')
-    responseMock = mocker.Mock()
-    responseMock.status_code = 200
-    responseMock.text = _text
-    mocker.patch(
-        'requests.Session.get').return_value.__enter__.return_value = responseMock
-
     chat = LiveChat(video_id='__test_id__', processor=DummyProcessor())
     chats = chat.get()
     rawdata = chats[0]["chatdata"]

tests/test_parser.py

@@ -1,21 +1,18 @@
 from pytchat.parser.live import Parser
 import json
-from aioresponses import aioresponses
 from pytchat.exceptions import NoContents
 
 
+parser = Parser(is_replay=False)
+
+
 def _open_file(path):
     with open(path, mode='r', encoding='utf-8') as f:
         return f.read()
 
 
-parser = Parser(is_replay=False)
-
-
-@aioresponses()
 def test_finishedlive(*mock):
     '''配信が終了した動画を正しく処理できるか'''
-
     _text = _open_file('tests/testdata/finished_live.json')
     _text = json.loads(_text)
 
@@ -26,10 +23,8 @@ def test_finishedlive(*mock):
         assert True
 
 
-@aioresponses()
 def test_parsejson(*mock):
     '''jsonを正常にパースできるか'''
-
     _text = _open_file('tests/testdata/paramgen_firstread.json')
     _text = json.loads(_text)
 

tests/test_videoinfo.py

@@ -1,5 +1,6 @@
+from json.decoder import JSONDecodeError
 from pytchat.tool.videoinfo import VideoInfo
-from pytchat.exceptions import InvalidVideoIdException
+from pytchat.exceptions import InvalidVideoIdException, PatternUnmatchError
 
 
 def _open_file(path):
@@ -12,13 +13,13 @@ def _set_test_data(filepath, mocker):
     response_mock = mocker.Mock()
     response_mock.status_code = 200
     response_mock.text = _text
-    mocker.patch('requests.get').return_value = response_mock
+    mocker.patch('httpx.get').return_value = response_mock
 
 
 def test_archived_page(mocker):
     _set_test_data('tests/testdata/videoinfo/archived_page.txt', mocker)
     info = VideoInfo('__test_id__')
-    actual_thumbnail_url =  'https://i.ytimg.com/vi/fzI9FNjXQ0o/hqdefault.jpg'
+    actual_thumbnail_url = 'https://i.ytimg.com/vi/fzI9FNjXQ0o/hqdefault.jpg'
     assert info.video_id == '__test_id__'
     assert info.get_channel_name() == 'GitHub'
     assert info.get_thumbnail() == actual_thumbnail_url
@@ -30,7 +31,7 @@ def test_archived_page(mocker):
 def test_live_page(mocker):
     _set_test_data('tests/testdata/videoinfo/live_page.txt', mocker)
     info = VideoInfo('__test_id__')
-    '''live page :duration = 0'''        
+    '''live page: duration==0'''
     assert info.get_duration() == 0
     assert info.video_id == '__test_id__'
     assert info.get_channel_name() == 'BGM channel'
@@ -64,3 +65,37 @@ def test_no_info(mocker):
     assert info.get_title() is None
     assert info.get_channel_id() is None
     assert info.get_duration() is None
+
+
+def test_collapsed_data(mocker):
+    '''Test case the video page's info is collapsed.'''
+    _set_test_data(
+        'tests/testdata/videoinfo/collapsed_page.txt', mocker)
+    try:
+        _ = VideoInfo('__test_id__')
+        assert False
+    except JSONDecodeError:
+        assert True
+
+
+def test_pattern_unmatch(mocker):
+    '''Test case the pattern for extraction is unmatched.'''
+    _set_test_data(
+        'tests/testdata/videoinfo/pattern_unmatch.txt', mocker)
+    try:
+        _ = VideoInfo('__test_id__')
+        assert False
+    except PatternUnmatchError:
+        assert True
+
+
+def test_extradata_handling(mocker):
+    '''Test case the extracted data are JSON lines.'''
+    _set_test_data(
+        'tests/testdata/videoinfo/extradata_page.txt', mocker)
+    try:
+        _ = VideoInfo('__test_id__')
+        assert True
+    except JSONDecodeError as e:
+        print(e.doc)
+        assert False